LmxMPK3, a mitogen-activated protein
kinase involved in length control of a
eukaryotic flagellum
DISSERTATION
submitted for the doctoral degree
- Dr. rer. nat. -
Department of Biology,
Faculty of Mathematics, Informatics and Natural Sciences,
University of Hamburg, Germany
by
Maja Erdmann
Hamburg, Germany
May 2009
We absolutely must leave room for doubt or there is no progress
and no learning. There is no learning without having to pose a
question. And a question requires doubt.
Table of contents
1 Introduction
1
1.1 Leishmania
and
leishmaniasis
1
1.1.1 Taxonomy of Leishmania species 1 1.1.2 Clinical manifestation and epidemiology of leishmaniases 1 1.1.3 Current anti-leishmanial chemotherapies 3
1.1.4 Leishmania life cycle 4
1.1.5 Genome organisation and gene regulation in Leishmania 7
1.2 The eukaryotic flagellum
10
1.2.1 Structure and function of the flagellum 10
1.2.2 Intraflagellar transport (IFT) 12
1.3 Signal transduction in eukaryotic cells
15
1.3.1 Different signalling pathways 15 1.3.2 Protein phosphorylation and protein kinases 17
1.3.3 The MAP kinase cascade 19
1.3.4 Signal transduction in trypanosomatids 20
1.4 LmxMPK3 - State of knowledge and project aims
26
1.4.1 State of knowledge 26
1.4.2 Project aims 27
2
Materials
29
2.1 Laboratory equipment
29
2.2 Plastic and glass wares, other materials
30
2.3
Chemicals
30
2.4 Culture media, stock and buffer solutions
33
2.5 Bacterial strains
37
2.6 Leishmania strains
38
2.7
Mouse
strain
38
2.8 DNA vectors and plasmid constructs
38
2.9
Oligonucleotides
39
2.10
Antibodies
40
2.11
Enzymes
40
2.12 Molecular biology kits
41
3
Methods
42
3.1
Cell
biology
methods
42
3.1.1 Culturing of E. coli 42
3.1.1.1 Culturing on medium plates 42
3.1.1.2 Culturing in liquid medium 42
3.1.1.3 Preparation of glycerol stocks 42
3.1.2 Culturing of Leishmania 42
3.1.2.1 Culturing of L. mexicana and L. major promastigotes 42 3.1.2.2 In vitro differentiation to L. mexicana axenic amastigotes 42 3.1.2.3 In vitro differentiation to L. mexicana promastigotes 43 3.1.2.4 Preparation of Leishmania stabilates 43
3.1.3 Leishmania cell counting 43
3.1.4 Fluorescence-activated cell sorting (FACS) of Leishmania promastigotes 43
3.2
Molecular
biology
methods
43
3.2.1 Preparation of competent bacteria 43
3.2.1.1 Method of Hanahan (1983) 43
3.2.1.2 Preparation of competent BL21 (DE3) [pAPlacIQ] 44
3.2.2 Transformation of E. coli 44
3.2.3 Transfection of Leishmania 44
3.2.3.1 Gene Pulser transfection (BIO RAD) 44 3.2.3.2 Nucleofector transfection (Amaxa) 45 3.2.4 Isolation of plasmid DNA from E.coli 45 3.2.4.1 Plasmid DNA mini-preparation (Zhou et al., 1990) 45 3.2.4.2 Plasmid DNA mini-preparation using Macherey & Nagel and Qiagen Kits 46 3.2.4.3 Plasmid DNA midi-preparation using Invitrogen, Macherey & Nagel and 46
Qiagen Kits
3.2.5 Isolation of genomic DNA from Leishmania (Medina-Acosta and Cross, 1993) 46 3.2.6 Isolation of total RNA from Leishmania using a Macherey & Nagel Kit 47 3.2.7 Determination of DNA and RNA concentrations 47 3.2.8 Reactions with DNA-modifying enzymes 47 3.2.8.1 Cleavage of DNA using type II restriction endonucleases 47 3.2.8.2 Complete fill-in of a 5’-overhang to create blunt end DNA using Klenow 47 enzyme
3.2.8.3 Dephosphorylation of DNA 5’-ends 47 3.2.8.4 Ligation of DNA fragments 48 3.2.9 Phenol/chloroform extraction of aqueous DNA solutions 48
3.2.10 Ethanol precipitation of DNA 48
3.2.12 DNA extraction from agarose gels using Macherey & Nagel and Qiagen Kits 48 3.2.13 Insertion mutagenesis using complementary 5’-phosphorylated 49 oligonucleotides
3.2.14 Polymerase chain reaction (PCR) 49 3.2.15 Reverse transcription-polymerase chain reaction (RT-PCR) 50 3.2.16 Cloning of a PCR product using the TOPO TA Cloning Kit (Invitrogen) 50
3.2.17 DNA sequencing 50
3.2.18 Southern blot analysis 50
3.2.18.1 Cleavage of genomic DNA and agarose gel electrophoresis 50 3.2.18.2 Denaturation, capillary blotting and cross-linking of DNA to nylon 51 membrane
3.2.18.3 Pre-hybridisation, hybridisation and stringency washing 51 3.2.18.4 Detection of the DIG-labelled hybridisation probe 51 3.2.18.5 Stripping-off the hybridisation probe 52
3.3 Protein and immunochemical
methods
52
3.3.1 Expression of recombinant proteins in E. coli 52 3.3.2 Preparation of E. coli cell lysates for protein purification 52 3.3.3 Affinity purification of recombinant proteins 53 3.3.3.1 Purification of GST-fusion proteins 53 3.3.3.2 Purification of His-tag fusion proteins 53 3.3.4 Thrombin cleavage of a GST-fusion protein to remove the GST-tag 53 3.3.5 Phosphoprotein purification from Leishmania using a Qiagen Kit 53 3.3.6 Determination of protein concentrations using Bradford reagent 54 3.3.7 Preparation of Leishmania lysates for immunoblot analysis 54 3.3.8 Preparation of Leishmania S-100 lysates for in vitro kinase assays 54 3.3.9 Discontinuous SDS polyacrylamide gel electrophoresis (SDS-PAGE) 54
3.3.10 Staining of SDS-PA gels 54
3.3.10.1 Coomassie staining 54
3.3.10.2 Silver staining 55
3.3.11 Drying of SDS-PA gels 55
3.3.12 Immunoblot analysis 55
3.3.12.1 Electroblotting of proteins using the semi-dry method 55 3.3.12.2 Immunological detection of proteins 56 3.3.12.3 Stripping-off the antibodies 56
3.4 In vitro kinase assays
56
3.4.1 Standard kinase assay with recombinant proteins 56 3.4.2 In vitro activation of recombinant kinases 56
3.4.3 Kinase assays with an activated recombinant kinase on Leishmania S-100 57 lysates
3.5 Mouse foot pad infection studies
57
3.6 Isolation of Leishmania amastigotes from mouse lesions
57
3.7 Microscopy techniques and flagellar length determination
58
3.7.1 Immunofluorescence analysis 58
3.7.2 Fluorescence microscopy on living Leishmania promastigotes 58 3.7.3 Transmission electron microscopy 59
3.7.4 Flagellar length determination 59
4
Results
60
4.1 The phenotype of the LmxMPK3 null mutants and the LmxMPK3 add
60
back mutants
4.1.1 Generation of the LmxMPK3 add back mutants 60 4.1.2 LmxMPK3 expression levels of the LmxMPK3 mutants 60 4.1.3 Measurements of the flagellar lengths of the LmxMPK3 mutants 61 4.1.4 Analysis of the ultrastructure using transmission electron microscopy 62 4.1.5 Quantification of PFR-2 in the LmxMPK3 null mutants 64 4.1.5.1 Immunofluorescence analysis 64
4.1.5.2 Immunoblot analysis 64
4.1.6 Mouse infection studies with the LmxMPK3 mutants 65
4.2 The expression profile of LmxMPK3 during differentiation of L. mexicana 66
4.3 Generation and characterisation of a GFP-LmxMPK3 and a GFP-
67
LmxMKK mutant
4.3.1 Preparation of the different transfection constructs 67 4.3.2 Transfection and verification of obtained clones 68 4.3.3 Measurements of the flagellar lengths of the GFP-LmxMPK3 and GFP- 70 LmxMKK mutants
4.3.4 Localisation studies of LmxMPK3 and LmxMKK using fluorescence 71 microscopy on living cells
4.3.5 Determining the correlation between LmxMPK3 amount and flagellar 72 length using fluorescence-activated cell sorting
4.4 Generation and characterisation of an inhibitor-sensitised LmxMPK3
73
mutant
4.4.1 Preparation of the transfection construct 74 4.4.2 Transfection and verification of obtained clones 74 4.4.3 Measurements of the flagellar lengths of the inhibitor-sensitised LmxMPK3 75 mutants
4.5 Biochemical characterisation of GST-LmxMPK3 and GST-LmxMPK3-KM 78
4.5.1 Generation of the expression constructs 78 4.5.2 Recombinant expression and affinity purification of GST-LmxMPK3 and 78 GST-LmxMPK3-KM4.5.3 Optimisation of the kinase assay reaction conditions for GST-LmxMPK3 79 4.5.4 Kinase assays with GST-LmxMPK3 and GST-LmxMPK3-KM 80
4.6 Analysis and optimisation of the activation of LmxMPK3 and LmxMPK3-
81
KM by LmxMKK-D
4.6.1 Kinase assays with in vitro-activated GST-LmxMPK3 and GST-LmxMPK3-KM 81 4.6.2 Optimisation of the LmxMPK3 activation using an in vivo system 82 4.6.2.1 Generation of the co-expression constructs 82 4.6.2.2 Recombinant co-expression and affinity purification of His-LmxMPK3 83 and His-LmxMPK3-KM
4.6.2.3 Optimisation of the kinase assay reaction conditions for in vivo-activated 84 His-LmxMPK3
4.6.2.4 Kinase assays with His-LmxMPK3 and His-LmxMPK3-KM derived from 85 the different co-expressions
4.7 Analysis of the activation mechanism of LmxMPK3
86
4.7.1 In vitro studies 87
4.7.1.1 Generation of the co-expression constructs 87 4.7.1.2 Recombinant co-expression and affinity purification of the different His- 88 LmxMPK3-TDY mutants
4.7.1.3 Kinase assays and subsequent analysis of the tyrosine phosphorylation 89 state of the different His-LmxMPK3-TDY mutants
4.7.1.4 Analysis of the phosphorylation state of the different His-LmxMPK3-TDY 92 mutants by mass spectrometry
4.7.2 In vivo studies 93
4.7.2.1 Generation of the different transfection constructs 93 4.7.2.2 Transfection and verification of obtained clones 95 4.7.2.3 Measurements of the flagellar lengths of the LmxMPK3-TDY mutants 98
4.8 Substrate search for LmxMPK3
100
4.8.1 PFR-2 as a potential LmxMPK3 substrate 101 4.8.1.1 Immunoblot analysis of PFR-2 in L. mexicana phosphoprotein fractions 101 4.8.2 A PFR-2 mRNA regulating protein as a potential LmxMPK3 substrate 102 4.8.2.1 RT-PCR analysis of PFR-2 mRNA in LmxMPK3 null mutants 102 4.8.3 An OSM3-like kinesin as a potential LmxMPK3 substrate 103 4.8.3.1 Generation of the expression construct 104 4.8.3.2 Recombinant expression and affinity purification of GST-LmxKin32 104 4.8.3.3 Kinase assays with GST-LmxKin32 and in vitro-activated GST-LmxMPK3 105
4.8.4 The outer dynein arm docking complex (ODA-DC) subunit DC2 as a potential 106 LmxMPK3 substrate
4.8.4.1 Immunoblot analysis of LmxDC2 in L. mexicana phosphoprotein fractions 106 4.8.4.2 Kinase assays with His-LdDC2 and in vivo-activated His-LmxMPK3 107 4.8.5 Glutamine synthetase as a potential LmxMPK3 substrate 108 4.8.5.1 Immunoblot analysis of LmxGS in L. mexicana phosphoprotein fractions 108 4.8.6 Kinase assays with in vitro-activated GST-LmxMPK3 on Leishmania lysates 109 4.8.7 In silico substrate search for LmxMPK3 using PREDIKIN 111 4.8.7.1 Features of LmjHS and its homologues 112 4.8.7.2 Testing the predicted LmjHS peptide as an LmxMPK3 substrate in vitro 113 4.8.7.3 Testing an N-terminal part of LmjHS as a LmxMPK3 substrate in vitro 116 4.8.7.4 Testing an N-terminal part of LmxHS as a LmxMPK3 substrate in vitro 118
5
Discussion
122
5.1 The phenotype of the LmxMPK3 null mutants and the LmxMPK3 add 122
back mutants
5.1.1 The morphology and structure of the flagellum 122 5.1.2 The ability to complete the life cycle 127
5.2 The expression profile of LmxMPK3 during differentiation of L. mexicana 128
5.3 The subcellular localisation of LmxMPK3 and its activator LmxMKK
129
5.4 Characterisation of an inhibitor-sensitised LmxMPK3 mutant - an inducible 131
system for selective kinase silencing
5.5 The correlation between LmxMPK3 amount and activity, and flagella length 133
5.6 Biochemical characterisation of LmxMPK3 and LmxMPK3-KM
134
5.7 The activation of LmxMPK3 and its molecular mechanism
135
5.7.1 Phosphorylation and activation of LmxMPK3 and LmxMPK3-KM 135 5.7.2 Phosphorylation and activation of different LmxMPK3-TDY mutants 1385.7.2.1 In vitro studies 139
5.7.2.2 In vivo studies 142
5.8 Substrate search for LmxMPK3
144
5.8.1 Testing potential candidate proteins for LmxMPK3 substrate function 145 5.8.1.1 PFR-2 and a PFR-2 mRNA regulating protein 145 5.8.1.2 The OSM3-like kinesin LmxKin32 146 5.8.1.3 The outer dynein arm docking complex (ODA-DC) subunit DC2 148 5.8.1.4 Glutamine synthetase 148 5.8.2 Screening the entire Leishmania proteome for LmxMPK3 substrates 149 5.8.2.1 In vitro kinase assays with activated LmxMPK3 on Leishmania lysates 149 5.8.2.2 In silico substrate search for LmxMPK3 using PREDIKIN 150
5.9 LmxMPK3 as a target for blocking leishmanial transmission to the insect 155
vector
5.10 LmxMPK3 mutants as model systems to study human ciliopathies
156
6
Summary
159
7
References
162
8
Appendix
178
8.1 Nucleotide and amino acid sequences
178
8.1.1 LmxMPK3 178 8.1.2 PFR-2C 181 8.1.3 LmxKin32 182 8.1.4 LmjDC2 183 8.1.5 LmxGS 184 8.1.6 LmjHS and LmxHS 184 8.1.7 IFT57 188
8.2
Plasmid
maps
189
Abbreviations
-/- double-allele deletion +/- single-allele deletion °C degree Celsius 1-NA-PP1 1-naphthyl-pyrazolo[3,4d]pyrimidine A ampère aa amino acidsADP adenosine diphosphate Amp ampicillin
APS ammonium persulfate ARE AU-rich element
ATP adenosine triphosphate BBS Bardet-Biedl syndrome
BLE phleomycin resistance marker gene
BNI Bernhard Nocht Institute for tropical medicine bp base pairs
BSA bovine serum albumine
C. elegans Caenorhabditis elegans
C. reinhardtii Chlamydomonas reinhardtii
CaBP Ca2+-binding proteins
cAMP cyclic adenosine monophosphate CD-domain common docking domain
cDNA complementary DNA
cGMP cyclic guanosine monophosphate CL cutaneous leishmaniasis CPB cysteine protease B
CSPD disodium 3-(4-methoxyspiro {1,2-dioxetane-3,2-(5-chloro)tricycle [3.3.1.13,7]decan}-4-yl)phenyl phosphate
Da Dalton
DABCO 1,4-diazabicyclo[2.2.2]octane DAG diacylglycerol
DAPI 4′,6-diamidino-2-phenylindole dilactate
DB database
DCL diffuse cutaneous leishmaniasis ddH2O double distilled water
D-domain docking domain DGC directional gene cluster
DHFR-TS dihydrofolate reductase-thymidylate synthase DIC differential interference contrast
DIG digoxigenin
DMF N,N-dimethylformamide DMSO dimethyl sulfoxide
DNA deoxyribonucleic acid
dNTP deoxyribonucleotide triphosphate DTT 1,4-dithiothreitol
E. coli Escherichia coli
EDTA ethylenediamine tetraacetic acid EGF epidermal growth factor
EGTA ethylene glycol bis(β-aminoethylether) tetraacetic acid EPB electroporation buffer
ER endoplasmic reticulum
ERK extracellular signal-related kinase F Farad
FACS fluorescence-activated cell sorting FAZ flagellar attachment zone
FCaBP flagellar Ca2+-binding protein
FCS fetal calf serum FML fucose mannose ligand g gramme
× g times gravity gDNA genomic DNA
GFP green fluorescent protein GIPL glycoinositol phospholipids gRNA guide RNA
GS glutamine synthetase GSK glycogen synthase kinase GST glutathione-S-transferase GTP guanosine triphosphate h hours
HEPES N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid His histidine
HPLC high performance liquid chromatography HRE hormone response element
HS hypothetical substrate HSP heat-shock protein
HYG hygromycin B resistance marker gene
IF immunofluorescence iFCS heat-inactivated FCS IFT intraflagellar transport IgG immunoglobulin G
iNOS inducible nitric oxide synthase InsP Inositol phosphate
InsP3 inositol 1,4,5-triphosphate
IPS myo-inositol-1-phosphate synthase
IPTG isopropyl-β-D-thiogalactopyranoside IR intergenic region
JNK c-Jun N-terminal kinase kb kilo base pairs
kDa kilo Dalton kDNA kinetoplast DNA l litres L. Leishmania LB lysogeny broth LPG lipophosphoglycan M molar m/z mass-to-charge ratio
MALDI-TOF matrix-assisted laser desorption/ionisation - time of flight MAP mitogen-activated protein
MAPK MAP kinase
MAPKAPK MAPK-activated protein kinase MAPKK MAP kinase kinase
MAPKKK MAP kinase kinase kinase MBP myelin basic protein
MCL mucocutaneous leishmaniasis MCS multiple cloning sites
MDA mass drug administration MES morpholinoethane sulfonic acid min minutes
MOPS morpholinopropane sulfonic acid mRNA messenger RNA
MS mass spectrometry
MS/MS tandem MS
NEO neomycin resistance marker gene
OD optical density
ODA-DC outer dynein arm docking complex ORF open reading frame
PAC puromycin resistance marker gene
PBS phosphate-buffered saline PCR polymerase chain reaction PFR paraflagellar rod
PH pleckstrin homology PhD Philosophiae Doctor
PKA protein kinase A
PKD polycystic kidney disease
PKDL post kala azar dermal leishmaniasis PM peritrophic membrane
PMSF phenylmethyl sulfonyl fluoride PSG promastigote secretory gel PTB phosphotyrosine binding PtdIns phosphatidylinositol
PtdInsP phosphatidylinositol phosphate PV parasitophorous vacuoles PVDF polyvinylidene fluoride RNA ribonucleic acid RNAi RNA interference RP retinitis pigmentosa rpm revolutions per minute rRNA ribosomal RNA
RT reverse transcriptase RT room temperature RTK receptor tyrosine kinases
RT-PCR reverse transcription-polymerase chain reaction s seconds
SAP shrimp alkaline phosphatase SDR substrate-determining residue SDS sodium dodecyl sulphate
SDS-PAGE SDS-PA gel electrophoresis SH Src homology
SL spliced leader SSC standard saline citrate
T. Trypanosoma
TBS Tris-buffered saline
TBV transmission-blocking vaccines TEM transmission electron microscopy TEMED N,N,N′,N′-tetramethylethylenediamine
TLCK Nα-tosyl-L-lysine chloromethyl ketone hydrochloride
TPR tetratricopeptide repeat
Tris tris(hydroxymethyl)aminomethane TRP transient receptor potential
U units
UTR untranslated region UV ultraviolet
V volt
v/v volume per volume VL visceral leishmaniasis w/v weight per volume
WHO World Health Organisation
1 Introduction
1.1 Leishmania and leishmaniasis
Leishmania are parasitic protozoa and the etiological agents of the leishmaniases, a group of
diseases transmitted to mammals by sand flies. The parasites were first discovered in India by the British tropical physician Sir W.B. Leishman and the Irish tropical physician C. Donovan in 1901. Leishmaniasis belongs to the currently 14 neglected tropical diseases listed by the World Health Organization (WHO).
1.1.1 Taxonomy of Leishmania species
Leishmania belong to the order Kinetoplastida named for the presence of the kinetoplast, a
distinct region of the single mitochondrion containing coiled DNA filaments, which is always closely associated with the basal body of the flagellum. Kinetoplastida can be divided into two suborders according to the number of flagella per cell. While Bodonina reveal two flagella and are mostly free-living, Trypanosomatina possess only a single flagellum and are predominantly parasitic. The latter comprise the family Trypanosomatidae consisting of nine different genera. While some of them use plants, insects or reptiles as their main hosts, the genera Endotrypanum, Trypanosoma and Leishmania infect mammals. Besides
Trypanosoma brucei (gambiense and rhodesiense) causing sleeping sickness and
Trypanosoma cruzi causing Chagas disease, also 21 of the almost 30 known Leishmania
species are pathogen to humans.
1.1.2 Clinical manifestation and epidemiology of leishmaniases
Since Leishmania parasites are transmitted to mammals by sand flies, their endemic region is consistent with the habitat of their vectors, predominantly rural areas in the tropics and subtropics. Sand flies of the genus Phlebotomus are found in the old world (Africa, Asia and Europe), whereas the genus Lutzomyia is found in the new world (America). Both genera belong to the family of Psychodidae (moth fly) in the order of Diptera. Leishmaniasis occurs in 88 countries worldwide distributed on all continents except Australia. 72 of those nations belong to the developing countries including the 13 poorest countries in the world. However, leishmaniasis is also found in 16 European countries like France, Spain, Italy and Greece. Over the last 10 years endemic regions have been spreading, and a significant increase in the number of recorded cases of the disease has been reported. The WHO estimates that 2 million new cases occur annually, 12 million people are presently infected and 350 million people are currently threatened by the disease worldwide.
Figure 1: Geographical distribution of leishmaniases
A: cutaneous and mucocutaneous leishmaniasis; B: visceral leishmaniasis. (Source: http://www.infektionsbiologie.ch/modellparasiten/leishmania.htm, 2004)
There are three different forms of leishmaniasis differing in their clinical symptoms. The clinical manifestation and the severity of the disease depend on the Leishmania species as well as on the genotype and immune status of the host.
Cutaneous leishmaniasis (CL)
CL, also known as Aleppo boil, Bagdad boil or oriental boil, is the most common form of leishmaniasis which exclusively affects the skin. A lesion develops at the site of bite typically located on exposed areas such as the face, arms and legs and mostly remains restricted to this site. The lesion often spontaneously heals with scarring accompanied by a lifelong immunity against the Leishmania species which caused the disease. CL accounts for ca. 75% of new Leishmania infections. 90% of the cases of CL are found in Iran, Afghanistan, Syria, Saudi Arabia, Brazil and Peru. The CL-causing species in the old world are predominantly L. major, L. tropica and L. aethiopica. CL in the new world is mainly caused by members of the L. mexicana complex, L. panamensis and L. guyanensis.
A more severe form of CL is the diffuse cutaneous leishmaniasis (DCL) resulting in widely spread and chronic skin lesions which may cover an individual’s entire body. This form of CL is difficult to treat and patients do not self-cure. DCL is found in Africa and America and is caused by some members of the L. mexicana complex and L. aethiopica.
Mucocutaneous leishmaniasis (MCL)
MCL, also called Uta or Espundia, is exclusively found in America and is mostly caused by members of the L. braziliensis complex. This form of leishmaniasis can lead to an extensive destruction of the nasal, pharyngeal and laryngealmucosa and their surrounding tissues. It develops as a complication of CL with parasites disseminating from the primary cutaneous lesion via lymphatic and blood vessels to reach the upper respiratory tract mucosa. MCL is difficult to treat and can be fatal especially if superinfections occur.
Visceral leishmaniasis (VL)
VL is the most severe form of leishmaniasis and also known as kala azar or Dum-Dum fever. After infection the parasite migrates to the internal organs such as liver, spleen and bone
A
B
marrow. Typical symptoms include fever, weight loss, anaemia and substantial swelling of the liver and spleen. If left untreated the disease results in the death of the host within two years. VL is caused by members of the L. donovani and L. infantum complex. 90% of the cases of VL are found in Bangladesh, India, Nepal, Sudan and Brazil.
Patients who have recovered from VL may suffer from post kala azar dermal leishmaniasis (PKDL). PKDL is a chronic form of CL beginning with a nodular rash appearing on the face which then spreads to other parts of the body. The disease is particularly severe if the lesions spread to the mucosal surfaces.
Figure 2: Clinical picture of leishmaniases
A: CL; B: MCL; C: VL. (Source: WHO/TDR/Crump/Kuzoe)
1.1.3 Current anti-leishmanial chemotherapies
Since over 60 years pentavalent antimonials have been used to treat all forms of leishmaniasis, especially CL and VL. The primary mode of action has not been clarified to date. Pentostam® (sodium stibogluconate, GlaxoSmithKline) and Glucantime® (meglumine antimoniate, Aventis) are the most conventional products. A disadvantage of this “first line drug” is the long period of parenteral administration for 20 to 28 days as well as severe side affects such as heart and liver damages (Lee and Hasbun, 2003). Moreover, antimonial resistance has been observed in India since 1980. As a consequence, 65% of Indian patients are not responsive to antimonial treatment any more (Sundar, 2003).
Due to the increasing resistance to pentavalent antimonials several “second line drugs” have been recommended or newly developed. The diamidine pentamidine was introduced in 1952 and is used to treat all forms of leishmaniasis in cases of antimonial resistance. The primary mode of action is unclear to date. The use of pentamidine is mainly restricted by several serious side effects such as hyperkalaemia, renal and gastrointestinal dysfunctions, and also the development of an insulin-dependent diabetes mellitus.
Amphotericin B is a highly effective polyene antibiotic which is used for treatment of antimonial resistant VL and certain cases of MCL (Croft and Coombs, 2003). The compound is selective towards ergosterol which is the predominant sterol over cholesterol in
and kidney damages, and myocarditis have been observed. Therefore, numerous lipid formulations of Amphotericin B were developed in the 1980s showing a reduced toxicity. AmBisome® (Gilead Sciences), the liposomal formulation of Amphotericin B, was first shown to be effective against VL in 1991. However, high costs restrict its use as an anti-leishmanial drug.
In 2002 Miltefosine (hexadecylphosphocholine) was registered in India for treatment of VL and CL in cases of antimonial resistance. It was the first anti-leishmanial drug for oral administration. However, Miltefosine has exhibited teratogenicity and thus should not be administered to women of child-bearing age (Croft and Coombs, 2003). Moreover, there are concerns regarding the long half-life which might support drug resistance. Therefore, a combination therapy treatment is recommended.
The aminoglycoside antibiotic paromomycin was registered in 2006 to treat VL in India. It is also used to treat CL in topical or parenteral formulations. Occurring side effects are mostly relatively harmless.
Due to the risk of occurring drug resistances the development of new anti-leishmanial drugs is paramount. New drugs should be orally administerable, financially affordable, well tolerated by patients and should optimally aim at more than one target structure to counteract against the development of drug resistances.
1.1.4 Leishmania life cycle
Leishmania parasites undergo profound biochemical and morphological changes when
passing through their digenetic life cycle, whereby they survive in their sand fly vector as well as in their mammalian host. Different cell surface glycoconjugates play an important role in the survival strategy of the parasite. The insect-stage promastigotes are spindle-shaped cells, 10 to 20 µm in length, which possess a flagellum, reaching up to 20 µm in length, protruding from the flagellar pocket at the anterior end of the cell. In contrast, the amastigotes living in the phagocytes of their mammalian host are spherical-shaped cells, only 2 to 4 µm in diameter, which reveal only a very short flagellum limited to the flagellar pocket. Both forms of the parasite multiply by binary fission.
Promastigotes assume different morphological forms in the gut of the sand fly (see Table 1). At different stages lipophosphoglycan (LPG), the main cell surface glycoconjugate of promastigotes, binds to lectin receptors of the gut epithelium to prevent expulsion of the parasite during defecation. Procyclic promastigotes are present in the abdominal midgut of the female sand fly and develop from amastigotes within 48 h after the blood meal while still enclosed by the peritrophic membrane (PM) protecting the parasite from digestive enzymes (Pimenta et al., 1997). They are an oval-shaped, flagellated, slightly motile and replicative form of promastigotes. During the following 24 h the procyclic forms slow down their
replication and transform into the non-dividing, long, slender and strongly motile nectomonad promastigotes which escape from the PM by secretion of a chitinase (Schlein et al., 1991) to anchor themselves to the midgut epithelium. They subsequently migrate towards the anterior midgut until reaching the stomodeal valve located at the junction between midgut and foregut. By day four the nectomonad forms develop into leptomonad promastigotes, shorter forms of the parasite which initiate a second growth cycle resulting in a massive infection. Leptomonads produce the promastigote secretory gel (PSG) which blocks the anterior midgut and ensures the transmission of the parasite to the mammalian host at a later stage. After day five leptomonad promastigotes differentiate into mammalian-infective, non-dividing metacyclic promastigotes. Additionally, leaf-like haptomonad forms with short flagella are observed at the stomodeal valve forming a parasite plug. Directly before the next blood meal the PSG plug has to be regurgitated by the female sand fly, thereby transmitting the metacyclic promastigotes into the skin of the mammalian host. It is assumed that an induced enzymatic damage of the stomodeal valve, an occurrence of parasites in the salivary glands and an excretion of parasites from the anus of infected sand flies additionally contributes to the transmission of the parasite.
Morphological form Critiria Schematic illustration
Amastigote ovoid body, flagellum not visible
Procyclic promastigote BL 6.5 - 11.5 µm, flagellum < BL
Nectomonad promastigote BL ≥ 12 µm, flagellar length variable
Leptomonad promastigote BL 6.5 - 11.5 µm, flagellum ≥ BL
Haptomonad promastigote disc-like expansion of flagellar tip,
body form and flagellar length variable
Metacyclic promastigote BL ≤ 8 µm, flagellum > BL
Table 1: Morphological forms of L. mexicana BL, body length. (modified from Rogers et al., 2002)
In the skin of the mammalian host the metacyclics avoid the complement-mediated lysis using different strategies. On the one hand, the insertion of the lytic C5b-C9 complex into the promastigote membrane is effectively blocked (Puentes et al., 1990). On the other hand, a serine/threonine protein kinase (LPK-1) is secreted by the metacyclics which leads to an inactivation of C3, C5 and C9 (Hermoso et al. 1991). At the same time, however, the promastigotes depend on fixation of opsonic complement factors to enter the host macrophages. Thus, the leishmanial surface metalloprotease gp63, also called leishmanolysin, rapidly converts bound C3b into iC3b favouring phagocytic clearance rather than lytic clearance. The uptake by macrophages is mediated by the complement receptors
CR1 and CR3 (Brittingham et al., 1995; Rosenthal et al., 1996) as well as by mannosyl-fucosyl receptors and fibronectin receptors in the macrophage membrane which bind to LPG and gp63 (Kane and Mosser, 2000).
Figure 3: Leishmania life cycle (Source: http://www.dpd.cdc.gov/dpdx)
The Leishmania-filled phagosomes subsequently fuse with lysosomes to form phagolysosomes which are also termed parasitophorous vacuoles (PVs). The PVs contain acid hydrolases at a local pH of 4.7 to 5.2 and vary in number, size and shape according to the respective Leishmania species. While numerous L. mexicana and L. amazonensis parasites share one PV, each L. major and L. donovani parasite is individually located in a small PV (Antoine et al., 1998). Triggered by the low pH and the elevated temperature in the mammalian host, the metacyclic promastigotes differentiate into amastigotes within 2 to 5 days (Shapira et al., 1988; Zilberstein et al., 1991). Subsequent proliferation of the amastigotes eventually leads to lysis of the macrophage and infection of neighbouring macrophages with the released amastigotes.
An early response of macrophages to the infection with pathogens is the respiratory burst, a rapid release of reactive oxygen species such as the hyperoxide anion (O2-) and hydrogen
peroxide (H2O2), and the generation of nitric oxide (NO). Leishmania parasites evade those
defence mechanisms using different strategies. It was found that LPG reduces the production of O2- by inhibiting protein kinase C. In addition, gp63 was shown to be involved in
the suppression of the respiratory burst (Sørensen et al., 1994). In addition, it was found that the inducible nitric oxide synthase (iNOS) is inhibited in early infection by both LPG and glycoinositol phospholipids (GIPLs), the major constituents of the amastigote surface
(Proudfoot et al., 1995). Moreover, the release of pro-inflammatory cytokines by the macrophage is prevented by affecting the phosphorylation state of MAP (mitogen-activated protein) kinases such as p38 and ERK 1/2 (Martiny et al., 1999; Prive and Descoteaux, 2000; Junghae and Raynes, 2002).
Apart from macrophages, other cell types are able to phagocytose Leishmania parasites. During the early phase of infection parasites are taken up by dendritic cells (Caux et al., 1995) as well as polymorphonuclear neutrophil granulocytes. The latter are believed to act as host cells before macrophages are infected (Laufs et al., 2002; Laskay et al., 2003). Moreover, fibroblasts presumably serve as host cells for persisting Leishmania parasites in the clinically latent disease (Bogdan et al., 2000).
1.1.5 Genome organisation and gene regulation in Leishmania
Supported by the WHO the sequencing of the L. major genome was started in 1994 by the
Leishmania Genome Network and was finished in 2003 (Ivens et al., 2005). Thereafter, the
L. infantum and L. braziliensis genome sequencing was completed by the Sanger Institute
(Peacock et al., 2007). Currently, sequencing of the L. mexicana genome is in progress, and shotgun reads are already available on the website of the Sanger Institute.
The haploid genome of Leishmania comprises 3.2 to 5 × 107 base pairs (bp) which are
arranged on 34 (e. g. L. mexicana; Britto et al., 1998) to 36 chromosomes depending on the
Leishmania species. As the chromosomes do not condense during the mitotic cycle, the
number of chromosomes was determined by pulsed field gel electrophoresis (Stiles et al., 1999). Like in other eukaryotes, chromosomes in Leishmania reveal telomeric sequences at their ends (Myler et al., 1999), however, they lack typical centromeric sequences. Chromosomal sizes range from 0.3 to 2.8 × 107 bp in L. major. Number and size of
chromosomes can change rapidly, since repetitive DNA sequences (30% of the genome) can cause amplifications or deletions of DNA regions by homologous intramolecular recombination. In addition to the standard complement of chromosomes, Leishmania parasites can contain linear or circular multi-copy minichromosomes which can constitute 5 to 10% of the total cellular DNA. They can form spontaneously or as a response to drug selection or nutrient stress and are the result of the amplification of DNA regions (Beverley, 1991; Segovia, 1994) which is likely to occur by homologous intramolecular recombination supported by flanking repetitive sequences (Olmo et al., 1995; Grondin et al., 1996).
Leishmania organisms are predominantly diploid as indicated by the need of two consecutive
rounds of electroporation for the generation of null mutants (Cruz et al., 1991) and the presence of restriction site polymorphisms (Hendrickson et al., 1993). However, contrary to
T. brucei, genetic (sexual) exchange in Leishmania seems to be an infrequent feature
of sequence polymorphisms in the Leishmania genome is very low (< 0.1%), contrasting with the genomes of T. brucei and T. cruzi (Ivens et al., 2005). The G/C content of the Leishmania genome is noticeably high (57%; Alonso et al., 1992) compared to the mammalian genome (40 to 45%), especially in the third (wobble) position of the amino acid codons (ca. 85%). So far, 8370 protein-coding genes have been identified in the L. major genome (http://www.genedb.org/genedb/leish).
Apart from the genomic DNA (gDNA) of the nucleus, Leishmania organisms contain the kinetoplast DNA (kDNA) which is located in the single, large mitochondrion of the parasite and makes up 10 to 15% of the total cellular DNA. The kDNA consists of several thousand circular, non-supercoiled DNA molecules which are catenated to generate a highly condensed planar network. There are two types of circular DNA molecules. Minicircles are present in 5000 to 10000 non-identical copies per cell ranging from 0.5 to 2.8 kilo base pairs (kb). So far, their only known genetic function is to encode guide RNAs (gRNAs) which are involved in the editing of maxicircle transcripts (see below). The maxicircles exist in 25 to 50 identical copies per cell, and their size ranges from 20 to 39 kb. They encode rRNAs, mitochondrial proteins and a small number of gRNAs.
Generally, Leishmania genes do not contain introns, and hence cis-splicing mechanisms are not expected to occur. So far, only four genes subjected to cis-splicing have been identified in trypanosomatids, among them an RNA helicase (Ivens et al., 2005). Almost one third of the Leishmania protein-coding genes is clustered into families of related genes. While smaller families have most likely developed by tandem gene duplication, genes of larger families have multiple loci consisting of single genes and/or tandem arrays and often represent Leishmania-specific genes. Genes of highly expressed proteins such as α- and β-tubulins, flagellar proteins, heat shock proteins (HSPs), proteases, transporters and surface proteins are present in multiple copies. Those genes are often organised as direct tandem repeats which most likely serves as a mechanism to increase the abundance of the primary transcripts. Some correlation between the gene copy number and the intracellular protein concentration was demonstrated for some heat shock proteins in Leishmania promastigotes (Brandau et al., 1995; Hübel et al., 1995).
Although trypanosomatids reveal a range of chromatin-remodelling activities, the mechanisms regulating RNA polymerase II-directed transcription seem to differ strongly from those of other eukaryotes. The chromosomes are organised into directional gene clusters (DGCs) of tens to hundreds of genes with unrelated predicted functions which can reach up to 1259 kb in size (Ivens et al., 2005). Those clustered genes are co-transcribed thus generating a polycistronic pre-mRNA. A common spliced leader (SL) sequence of 39 nucleotides, also known as mini-exon, is subsequently attached to the 5’-end of all messages by a mechanism called trans-splicing. The SL is encoded separately by
approximately 200 gene copies which are predominantly organised in a tandem array. The 5’-end of the SL contains a 7-methylguanosine cap which is essential for the splicing reaction. Trans-splicing is controlled by polypyrimidine (CT) tracts in the 5’-UTR of genes and usually occurs at the first AG dinucleotide downstream of the CT tract. In Leishmania 3’-end polyadenylation eventually releasing monocistronic mRNA is coupled to trans-splicing of the downstream gene neighbour (Ullu et al., 1993). Polyadenylation occurs 100 to 500 nucleotides upstream of the splice-acceptor site (Stiles et al., 1999; Clayton, 2002). Therefore, unlike other eukaryotes, poly(A) site selection is not determined by consensus poly(A) signal sequences. The 5’- and 3’-UTR of Leishmania transcripts are mostly longer than those of other eukaryotes reaching up to 688 bp and 2973 bp, respectively. The initiation mechanism of RNA polymerase II-directed transcription has not been clarified to date. The only known RNA polymerase II promoter belongs to the SL gene and is located upstream of each SL gene copy (Saito et al., 1994). Several homologues of RNA polymerase II basal transcription factors have been identified, however, the majority of those factors is missing (Ivens et al., 2005). By contrast, the trypanosomatid genomes contain a noticeably high number of genes encoding proteins with potential RNA binding properties. Current knowledge strongly suggests that gene expression in trypanosomatids is primarily regulated on the posttranscriptional level, contrasting higher eukaryotes controlling gene expression mainly by regulating transcription. Transcript abundance depends on sequences in the 3’-UTR and the downstream intergenic region (IR) affecting mRNA processing and/or stability and is mediated by labile protein factors (Stiles et al., 1999). The 3’-UTR is also known to control translation efficiency. In some cases, posttranslational modifications affect intracellular protein amounts (Clayton, 1999).
Another characteristic of trypanosomatids is the extensive sequence modification of the mitochondrial transcripts by a process known as RNA editing. The genes on the maxicircles generate transcripts lacking numerous uracil (U) units. The gRNA (see above) serves as a template for the insertion (or less frequently the deletion) of uracil into the pre-mRNA recruiting different enzymes. The process of RNA editing is essential for converting the mitochondrial transcripts into mature mRNAs ready for translation.
There are different ways for the analysis and manipulation of genes in Leishmania organisms. The generation of null mutants can be achieved by sequentially replacing both alleles of the gene to be analysed by different resistance marker genes in two consecutive rounds of electroporation. Gene replacement occurs by the mechanism of homologous recombination. A different strategy of gene introduction is the addition of an expression vector carrying the gene of interest in addition to selected IR sequences and a resistance marker gene. In contrast to T. brucei, the mechanism of RNA interference (RNAi) has not been applied successfully to Leishmania to date. However, while most Leishmania species
lack essential components involved in this process (Robinson and Beverley, 2003),
L. braziliensis has been found to contain all components necessary for RNAi (Peacock et al.,
2007).
1.2 The eukaryotic flagellum
1.2.1 Structure and function of the flagellum
Flagella and cilia are eukaryotic organelles conserved from protists to mammals. They function in a variety of biological processes such as single cell movement, sensory reception and fluid movement in complex multicellular organisms. Flagella show the same construction as cilia, however, they are much longer. They typically project from the cell surface and are composed of a microtubule backbone (axoneme) surrounded by a membrane contiguous with the plasma membrane. According to the axonemal organisation of microtubule pairs two main ciliary types, namely “9+2” (motile) and “9+0” (primary, non-motile), have been defined. The “9+2” axoneme is composed of nine microtubule doublets surrounding a central pair of singlets which is absent in the “9+0” axoneme. However, this classical distinction is obsolete since organisation of microtubules can vary within one organelle as shown for the cilia of sensory neurons of Caenorhabditis elegans displaying middle segments composed of nine microtubule doublets and distal segments of nine microtubule singlets. In addition, the classification of “9+2” and “9+0” as motile or sensory is strongly simplified. Examples of motile primary cilia (e.g. in the renal epithelium; Ong and Wagner, 2005) as well as motile cilia/flagella with sensory roles (see below) have been reported. The axonemal microtubules are linked to numerous other proteins such as the inner and outer dynein arms responsible for flagellar beating, the radial spokes and the nexin bridges (see Figure 4). It was found that the axoneme of Chlamydomonas species is composed of at least 250 proteins (Piperno et
al., 1977).
Figure 4: Schematic illustration of a “9+2” axoneme
A unique feature of the flagellum in trypanosomatids is the presence of a lattice-like structure called the paraflagellar rod (PFR) which is attached to the axoneme and runs along the length of the flagellum (Gull, 1999). A PFR has so far been identified in three groups of protists: kinetoplastids, euglenoids and dinoflagellates. The PFR consists of a short proximal, an intermediate and a more developed distal domain (see Figure 5). Filaments of the intermediate domain link the proximal and distal domains which are both composed of plate-like structures stacked parallel to each other. Moreover, the proximal domain of the PFR is connected to the axonemal microtubule doublets 4 to 7 by fibres. Although the complete composition of the PFR is still unknown, two closely related proteins could be identified as the major components. Their homologues are PFR-1 and PFR-2 in Leishmania, PFR-C and PFR-A in T. brucei, and PAR-3 and PAR-2 in T. cruzi, respectively. The corresponding genes are organised in tandem arrays of several gene copies. Some other proteins have been shown to localise to the PFR. Among them are calmodulin (Ruben and Patton, 1986; Ridgley
et al., 2000) and some calflagins (flagellar calcium-binding proteins; Wu et al., 1994; Bastin
et al., 1999a). In addition, several proteins involved in the nucleotide metabolism, namely
adenylate kinases (Pullen et al., 2004; Ginger et al., 2005) and cAMP phosphodiesterases (Zoraghi and Seebeck, 2002; Oberholzer et al., 2007), have been identified as PFR-associated components. Several examples show that the PFR has an essential role in the motility of the parasite. PFR-2 null mutans of L. mexicana still display a residual PFR containing PFR-1 subunits, however, they reveal an approximately 4-fold reduced velocity of forward motility (Santrich et al., 1997). The flagellar beat pattern is altered showing a reduced wavelength and a decreased beat frequency. The impaired motility might result from a reduced elastic bending resistance of the flagella lacking most of the PFR. A more dramatic phenotype could be generated in T. brucei in which the PFR-A mRNA was ablated by RNAi (Bastin et al., 1998; Bastin et al., 1999a; Bastin et al., 2000). The PFR seemed to be disrupted resulting in paralysed cells which sedimented to the bottom of the tissue culture flask. Besides its function in cell motility the PFR might serve as a scaffold for regulatory and metabolic proteins of the flagellum (Ralston and Hill, 2008).
Figure 5: Schematic illustration of the PFR next to the axoneme
The trypanosomatid flagellum is involved in other biological activities than cell motility such as the attachment to host surfaces (see 1.1.4) and intracellular signalling. The latter might be supported by several identified flagellar proteins with a potential role in environmental sensing or intracellular signal transduction. In T. brucei the adenylate cyclase ESAG4 is exclusively found in the flagellar membrane (Paindavoine et al., 1992), however, neither its function nor the corresponding signalling pathway have been identified to date. In addition, an EF-hand flagellar Ca2+-binding protein (FCaBP) was found to associate with the flagellar membrane in a Ca2+-dependent manner in T. cruzi (Engman et al., 1989). An analogous
mechanism can be found in the plasma membrane of mammalian retinal rod cells where the EF-hand Ca2+-binding protein recoverin mediates signal transduction by changes in
intracellular Ca2+ levels (Dizhoor et al., 1991; Calvert et al., 1995). Another example of a
flagellum-specific receptor is the glucose transporter ISO1 which was analysed by Piper et
al. (1995). Its possible role in glucose sensing is supported by observations made on yeast
and human orthologues (Ozcan et al., 1996; Bandyopadhyay et al., 2000). An example for the presence of motile cilia with sensory roles in mammals are the cilia of the female reproductive tract in mice which contain transient receptor potential (TRP) channels involved in environmental sensing (Teilmann et al., 2005). The flagellum of trypanosomes is additionally involved in regulating cell size, shape, polarity and division (Kohl et al., 2003). Those functions are likely to be mediated by the flagellar attachment zone (FAZ), a structure which is exclusively found in trypanosomes and is probably involved in the attachment of the flagellum to the cell body. It is therefore unlikely that the unattached flagellum of Leishmania has similar functions.
The flagellum of trypanosomatids exits from a deep invagination of the plasma membrane at the anterior end of the cell known as the flagellar pocket. Its opening is surrounded by the “zone of adhesion” (Overath et al., 1997), a desmosome-like thickening which might prevent the flow of material into and out of the flagellar pocket. However, macromolecules have been shown to pass this border (Landfear and Ignatushchenko, 2001). The flagellar pocket is the only site of the whole cell where endocytosis and the secretion of proteins take place. Moreover, membrane proteins are first delivered to the flagellar pocket from where they are differentially targeted to different membrane domains (Bastin et al., 2000).
1.2.2 Intraflagellar transport (IFT)
Intraflagellar transport (IFT) is the motor-dependent bidirectional movement of IFT particles along the length of eukaryotic flagella and cilia. It is a highly conserved process essential for the construction and maintenance of those organelles and has been excessively studied in
Chlamydomonas. Although many proteins involved in IFT have yet to be identified (Haycraft
conserved among green algae, nematodes and vertebrates have been identified. Morphologically similar particles are also present in the flagella of trypanosomatids (Sherwin and Gull, 1989). Using an in silico approach twelve of the conserved protein subunits could be identified in Leishmania (Gouveia et al., 2007) while at least 10 IFT complex proteins were found in T. brucei (Briggs et al., 2004; Absalon et al., 2008). IFT complex subunits are rich in protein-protein interaction domains of the tryptophan-aspartic acid (WD)-40, tetratricopeptide repeat (TPR) protein and coiled coil families (Cole, 2003) which allow complex assembly as well as binding of cargo and motor proteins.
Anterograde movement of IFT complex B (from base to tip) is driven by a heterotrimeric motor protein complex of the kinesin-2 family, consisting of two heterodimerised kinesin motor subunits and an accessory subunit termed kinesin-2-associated protein (KAP) (Cole et
al., 1993; Wedaman et al., 1996). Retrograde movement of IFT complex A (from tip to base)
is driven by the cytoplasmic motor protein complex dynein 1b (Pazour et al., 1999; Porter et
al., 1999; Signor et al., 1999) consisting of at least two subunits, namely a dynein heavy
chain (DHC1b) and a light intermediate chain (D2LIC) (Cole, 2003; Perrone et al., 2003).
Figure 6: Schematic illustration of intraflagellar transport (IFT) (Source: Cole, 2003)
Cilia contain multiple kinesins in addition to heterotrimeric kinesin-2 (Fox et al., 1994). Analysis of the IFT in chemosensory cilia of C. elegans revealed that besides heterotrimeric kinesin-2, termed kinesin-II, a second kinesin-2 family member, known as OSM-3, drives anterograde IFT as a homodimeric complex (Signor et al., 1999; Snow et al., 2004). While kinesin-II and OSM-3 function together to assemble the middle segment of the axoneme composed of microtubule doublets (with each motor being able to work in the absence of the other motor), OSM-3 alone extends the distal end consisting of microtubule singlets. However, OSM-3 only extends distal singlets in some ciliary types, since it is also active in amphid wing cilia of C. elegans which only reveal microtubule doublets (Scholey, 2008). KIF17, a close relative of OSM-3, is known to target cyclic nucleotide-gated channels to mammalian primary cilia (Jenkins et al., 2006). In C. elegans the kinesin-3 family member KLP-6 was shown to be essential for ciliary targeting of polycystins forming mechanosensory
ion channels in the membranes of cilia on male-specific sensory neurons (Peden and Barr, 2005). The microtubule-depolymerising kinesin-13 is supposed to cooperate with the IFT machinery at the flagellar tip to control the length of the flagellum in Leishmania (Blaineau et
al., 2007) and Giardia (Dawson et al., 2007). It is generally assumed that accessory kinesins
such as OSM-3 confer cilia-specific functions. They might be involved in modulating IFT, target specific proteins to the organelle or function as stable ciliary components (Scholey, 2008). In Leishmania a putative Unc104-like kinesin as well as a kinesin-2 subunit have been identified as IFT-related factors (Gouveia et al., 2007). In addition, the L. major genome DB (Ivens et al., 2005) reveals two putative OSM-3-like kinesins.
IFT is essential for the delivery of large numbers of different cargo proteins to the flagellum and back to the cell body. Axonemal subunits are transported from the basal body region to the tip of the flagellum, where the axoneme is assembled (Johnson and Rosenbaum, 1992). Additionally, IFT is responsible for the delivery of flagellar matrix and membrane proteins to the flagellum, with the latter proposed to be moved in the plane of the flagellar membrane (Qin et al., 2005). Also IFT complex A proteins and the inactive dynein motor complex are delivered to the flagellar tip to be unloaded, thereby keeping up retrograde IFT (Absalon et
al., 2008). In trypanosomatids the PFR has to be assembled as a separate structure apart
from the axoneme. Construction of the PFR is dependent upon IFT (Kohl et al., 2003), and PFR subunits are attached primarily at the flagellar tip (Bastin et al., 1999b). In return, kinesins, IFT complex B proteins as well as used axoneme and PFR subunits are transported back to the cell body for recycling or degradation. Anterograde and retrograde IFT proceed simultaneously resulting in a continuous turnover of flagellar subunits at the distal tip of the flagellum. Flagellar length might therefore be regulated by shifting the ratio between the rates of assembly and disassembly (Stephens, 1997; Marshall and Rosenbaum, 2001; Song and Dentler, 2001).
Likewise in Chlamydomonas, IFT particles have been localised to the space between the flagellar membrane and the axoneme in trypanosomes (Bastin et al., 2000). Remarkably, IFT particles are preferentially transported along the axonemal microtubule doublets 3 and 4, or 7 and 8, and thus along both sides of the PFR (Absalon et al., 2008). Although several IFT complex proteins have been localised along the length of the flagellum, a significant proportion is found around the area of the basal body (Cole et al., 1998; Deane et al., 2001; Pedersen et al., 2005). However, IFT-like particles are absent from the transition zone of the basal body (Absalon et al., 2008). It has been shown that IFT proteins are docked onto the transition fibers running between the basal body and the membrane (Deane et al., 2001). Therefore, the transition fibers might act as a staging area for IFT particle formation where IFT could be involved in the selection of proteins entering the flagellum (Cole, 2003).
Moreover, IFT seems to play a role in signal transduction (Sloboda, 2005) probably by moving sensor molecules to the flagellar tip where they might become modified. Returning the modified sensor molecules to the cell body would provide the cell with information about the state of the flagellum or the environment outside the cell (Pazour and Rosenbaum, 2002a). Actually, 93 signal transduction proteins could be identified in purified flagella of
C. reinhardtii which include 21 protein kinases (Pazour et al., 2005).
Several observations suggest that protein kinases are critically involved in flagellar length control. The aurora protein kinase CALK (Pan et al., 2004), the glycogen synthase kinase GSK3β (Wilson and Lefebvre, 2004), the CDK (cyclin-dependent kinase)-related kinase LF2 (Tam et al., 2007), the NIMA-related kinase Cnk2p (Bradley and Quarmby, 2005) and the MAP (mitogen-activated protein) kinase LF4 (Berman et al., 2003) have been shown to control flagellar length in Chlamydomonas. Also the MAP kinase DYF-5 in C. elegans has a role in flagellar length regulation (Burghoorn et al., 2007). Furthermore, the MAP kinases LmxMPK3 (Erdmann, diploma thesis, 2004; Erdmann et al., 2006), LmxMPK9 (Bengs et al., 2005), LmxMPK13 (the homologue of LF4) and LmxMPK14 (Scholz, PhD thesis, 2008), as well as the MAP kinase kinases LmxMKK (Wiese et al., 2003a) and LmxPK4 (Kuhn, PhD thesis, 2004) have been shown to regulate flagellar length in L. mexicana. Actually, more than 80 phosphorylated flagellar components have been identified in Chlamydomonas (Tuxhorn et al., 1998).
1.3 Signal transduction in eukaryotic cells
1.3.1 Different signalling pathways
Cells have to sense their environment to adapt or react to changes outside the cell. This feature is essential for single cell organisms such as Leishmania parasites as well as for cells in tissues or organs of multicellular organisms.
When Leishmania parasites pass through their digenetic life cycle they have to undergo profound biochemical and morphological changes to survive in the sand fly vector or in the mammalian host and to prepare for the next phase of their life cycle. Although it is not clear how environmental signals are sensed and transmitted into the cell, it is very likely that signal transduction processes are critically involved. Protein kinases are likely candidates, since
Leishmania parasites reveal stage-specific changes in protein phosphorylation (Dell and
Engel, 1994).
Multicellular organisms are further dependent on the efficient communication between single cells which are sometimes separated by long distances. Thus, hormones have taken over the task as extracellular chemical messengers transporting a signal from one cell to another.
Hormones can bind to different cellular receptors either being integral membrane proteins, cytoplasmic or nuclear proteins.
Hydrophobic hormones such as steroid hormones can pass the plasma membrane by passive diffusion and generally bind to nuclear receptors, a class of ligand-activated transcription factors, initially located in the cytosol. The hormone receptor complexes are subsequently translocated into the nucleus where they bind to specific nucleotide sequences known as hormone response elements (HREs), thereby regulating the transcription of different genes. Also soluble gases such as carbon monoxide (CO) and nitric oxide (NO) are able to diffuse into the cell where they activate a guanylate cyclase producing cyclic guanosine monophosphate (cGMP) as an intracellular messenger.
Hydrophilic hormones such as adrenalin act as “first messengers” by binding to a cell surface receptor resulting in a conformational change of the cytoplasmic receptor domain which eventually leads to the production of an intracellular signalling molecule, termed “second messenger”. This messenger triggers the intracellular release of Ca2+, alters gene expression
or activates different enzymes, which finally leads to changes in the metabolism or the cytoskeleton of the cell. Since those signal transduction pathways involve ordered sequences of biochemical reactions, they are also referred to as signalling cascades.
There are three main classes of cell surface receptors inducing specific intracellular responses:
Ligand-gated ion channel receptors mediate the quickest responses to extracellular signalling molecules. Binding of the messenger initiates temporary opening of the channel which leads to a change of ion concentrations over the membrane. The ion flow itself relays the signal and thus no “second messenger” is needed. An example for this mechanism is found in the post-synaptic cell of a neural synapse.
Seven-helix receptors such as the adrenergic receptor possess seven membrane-spanning α-helices and act through heterotrimeric GTP (guanosine triphosphate)-binding proteins which can switch between an active and an inactive form, thereby serving as molecular switches. The adenylate cyclase cascade is initiated generating cyclic adenosine monophosphate (cAMP) as a “second messenger”. In addition, the phosphoinositol cascade can be activated releasing inositol 1,4,5-triphosphate (InsP3) and diacylglycerol (DAG) as
“second messengers”. Both signalling pathways eventually lead to an increase of cytosolic Ca2+ levels. Ca2+ itself functions as an important intracellular messenger.
The third group describes the cell surface receptors with tyrosine kinase activity such as the epidermal growth factor (EGF) receptor. Those receptors are either linked with non-receptor tyrosine kinases on the cytosolic side of the plasma membrane or possess a cytoplasmic tyrosine kinase domain themselves. The latter are referred to as receptor tyrosine kinases (RTKs). Binding of the respective ligand to an RTK generally induces an oligomerisation of
the monomeric receptors. This leads to a crosswise tyrosine phosphorylation of the cytoplasmic receptor domains and to the phosphorylation of other submembraneous proteins on specific tyrosine residues. The phosphorylated tyrosine residues of the RTKs are recognised by proteins containing SH (Src homology)2 or PTB (phosphotyrosine binding) domains. Those proteins are adapter proteins, enzymes or subunits of the cytoskeleton. In addition, they often contain SH3 domains which bind to proline-rich sequence motifs of further cytoplasmic proteins. Eventually, the small (monomeric) GTP-binding protein Ras, a key component and switchpoint of different signalling pathways, is activated. Among others, the MAP (mitogen-activated protein) kinase cascade can be activated which is achieved by binding of activated Ras to the MAPKKK (MAP kinase kinase kinase) Raf, thereby triggering a conformational change and thus its activation. The MAP kinase cascade culminates in the phosphorylation of different substrate proteins such as enzymes and transcription factors. Signal silencing is an important feature of intracellular signalling mainly occurring through an inactivation of intracellular signalling components. Another characteristic is the organisation of individual pathways into complex networks leading to a “cross talk” between different signalling cascades. Thus, signals can be spread out to further pathways or can be joined and integrated resulting either in the amplification or in the attenuation of the signal (Frost et
al., 1997; Ganiatsas et al., 1998; Pearson et al., 2001; Sundaram, 2006).
1.3.2 Protein phosphorylation and protein kinases
Intra- and extracellular signals often lead to a change in the phosphorylation state of specific proteins in order to regulate important molecular processes within the cell. Phosphorylation is the most frequent posttranslational modification, with approximately one third of mammalian proteins being phosphorylated. Reversible phosphorylation can change the properties of a protein in many different ways by forming ionic and hydrogen bonds. Conformational changes as well as the generation or masking of binding motifs can result in an alteration of the enzymatic activity, protein stability, binding properties or the subcellular localisation. Protein kinases represent one of the largest gene families constituting 2% of all known mammalian genes. Those enzymes catalyse the transfer of the γ-phosphoryl group of adenosine triphosphate (ATP) or GTP to a hydroxyl group in their substrates. Most protein kinases belong to one of the three main groups named according to the amino acid residues being phosphorylated. While serine/threonine kinases are active toward serine and threonine residues, tyrosine kinases are highly specific for the phosphorylation of tyrosine residues. Dual-specificity kinases act on both aliphatic and aromatic amino acid residues. The latter group is only composed of MAPKKs (MAP kinase kinases) and LAMMER kinases (Hanks et
serine, threonine to tyrosine with a ratio of 1800:200:1 in eukaryotic cells (Hubbard and Cohen, 1993).
The structure of protein kinases is highly conserved and is composed of two domains flanking the catalytic cleft where ATP (or GTP) and the substrate can bind (see Figure 7). The smaller N-terminal lobe consists of a five-stranded antiparallel β-sheet (β1-β5) and an α-helix (αC) where the latter is involved in the orientation of the nucleotide substrate. The larger C-terminal lobe mostly consists of α-helices and is responsible for binding of the substrate and transferring the γ-phosphoryl group of ATP (or GTP) to a hydroxyl group in the substrate. The catalytic domain of protein kinases comprises approximately 300 amino acid residues and reveals twelve conserved subdomains separated by regions of lower homology (see Figure 8). While subdomains I to IV are located in the N-terminal lobe, subdomain V is found within the deep catalytic cleft and subdomains VI to XI in the C-terminal lobe.
Figure 7: 3D structure of the catalytic domain of a protein kinase
a: phosphate anchor ribbon; b: Lys-Glu ionic bond; c: catalytic loop; d: catalytic Asp of subdomain VIb; e: activation loop. (Source: Krupa et al., 2004)
Several conserved residues or secondary structures in both domains of protein kinases contribute to the orientation of the nucleotide substrate. Involved are the phosphate anchor ribbon which is a glycine-rich loop located between the β1- and β2-strands in subdomain I, an asparagine and an aspartate residue in subdomains VIb and VII, respectively, which bind divalent cations involved in nucleotide recognition, and a lysine residue in subdomain II which forms an ionic bond with a glutamate residue in the αC-helix and coordinates the α- and β-phosphoryl groups of the nucleotide substrate. The lysine residue is essential for the transfer of the γ-phosphoryl group of ATP. The so-called activation loop, typically 20-30 residues in length, provides a platform for the peptide substrate to bind in an extended conformation close to the γ-phosphoryl group of ATP. However, the activation loop has to be phosphorylated to be stabilised in an open and extended conformation which allows substrate binding and catalysis (Hubbard, 1997). The aspartate residue mentioned above is part of the highly conserved DFG motif located at the base of the activation loop. The
